The proposed study involves secondary analysis of existing clinical, administrative, and survey data sets. Capture- recapture analysis is used to create ascertainment corrected estimators of the prevalence of alcoholism, using data routinely gathered and computerized by an HMO. These estimators should facilitate health services planning by enabling health service providers to monitor changes in the prevalence of alcohol problems in a timely, cost-effective fashion. In addition, the proposed analysis will provides estimates of how much nonresponse and screening based on drinking behavior bias cross-sectional survey prevalence estimates. The ascertainment corrected estimators will be used to explore whether prevalence within this HMO has changed in the past five years and how demographic and health factors affect the prevalence of clinically identified alcoholism.